Constraint Aggregation in Infinite-Dimensional Spaces and Applications
Arkadii V. Kryazhimskii,
Andrzej Ruszczy
ski
Mathematical Steklov Institute, Russian Academy of Sciences, Gubkina 8, GSP-1, 117966 Moscow, Russia
RUTCOR and Department of Management Science and Information Systems, Rutgers University, Piscataway, New Jersey 08854
rusz{at}rutcor.rutgers.edu
An aggregation technique for constraints with values in Hilbert spaces is suggested. The technique allows us to replace the original optimization problem by a sequence of subproblems having scalar or finite-dimensional constraints. Applications to optimal control, games, and stochastic programming are discussed in detail.
Key Words: Constrained optimization in vector spaces; aggregation; optimal control; games; stochastic programming
History: Received: October 2, 1999;
revision received: June 9, 2000;
Copyright © 2001 by INFORMS.